Characterization of Microvessels and Parenchyma in In-Line Phase Contrast Imaging CT: Healthy Liver, Cirrhosis and Hepatocellular Carcinoma.
Jinghao Duan,Chunhong Hu,Qingtao Qiu,Jing Zhang,Huipeng Meng,Keqiang Wang,Huajiang Dong,Hong Wei,Yong Yin
DOI: https://doi.org/10.21037/qims.2019.06.12
2019-01-01
Quantitative Imaging in Medicine and Surgery
Abstract:Background: Hepatocellular carcinoma (HCC) is a cancer with a poor prognosis, and approximately 80% of HCC cases develop from cirrhosis. Imaging techniques in the clinic seem to be insufficient for revealing the microstructures of liver disease. In recent years, phase contrast imaging CT (PCI-CT) has opened new avenues for biomedical applications owing to its unprecedented spatial and contrast resolution. The aim of this study was to present three-dimensional (3D) visualization of human healthy liver, cirrhosis and HCC using a PCI-CT technique called in-line phase contrast imaging CT (ILPCI-CT) and to quantitatively evaluate the variations of these tissues, focusing on the liver parenchyma and microvasculature. Methods: Tissue samples from 9 surgical specimens of normal liver (n=3), cirrhotic liver (n=2), and HCC (n=4) were imaged using ILPCI-CT at the Shanghai Synchrotron Radiation Facility (SSRF) without contrast agents. 3D visualization of all ex vivo liver samples are presented. To quantitatively evaluate the vessel features, the vessel branch angles of each sample were clearly depicted. Additionally, radiomic features of the liver parenchyma extracted from the 3D images were measured. To evaluate the stability of the features, the percent coefficient of variation (%COV) was calculated for each radiomic feature. A %COV <30 was considered to be low variation. Finally, one-way ANOVA, followed by Tukey's test, was used to determine significant changes among the different liver specimens. Results: ILPCI-CT allows for a clearer view of the architecture of the vessels and reveals more structural details than does conventional radiography. Combined with the 31) visualization technique, ILPCI-CT enables the acquisition of an accurate description of the 3D vessel morphology in liver samples. Qualitative descriptions and quantitative assessment of microvessels demonstrated clear differences among human healthy liver, cirrhotic liver and HCC. In total, 38 (approximately 51%) radiomic features had low variation, including 11 first-order features, 16 GLOM features, 6 GLRLM features and 5 GLSZM features. The differences in the mean vessel branch angles and 3 radiomic features (first-order entropy, GLOM-inverse variance and GLCM-sum entropy) were statistically significant among the three groups of samples. Conclusions: ILPCI-CT may allow for morphologic descriptions and quantitative evaluation of vessel microstructures and parenchyma in human healthy liver, cirrhotic liver and HCC. Vessel branch angles and radiomic features extracted from liver parenchyma images can be used to distinguish the three kinds of liver tissues.